CxOPack
KIT · 01 / 05

CEO Kit

Strategy, decisions, leadership — in 15 minutes a week.

Your CEO Kit replaces what a real chief executive spends most of their time doing: setting priorities, making decisions, and communicating with investors. Use it as a weekly operating system, not as a chatbot.

#1. Review, plan, brainstorm

Before you install anything, write down answers to these — in Notion, a text file, anywhere. The kit is useful in direct proportion to how precisely you can answer these four questions.

Answer these first

  1. What is the single metric that will determine if this quarter was a success? (ARR, paying users, demos booked — pick one)
  2. If an investor asked you today 'what's the one thing that shipped this month that changes the trajectory?' — what would you say?
  3. What's the biggest decision that has been sitting on your desk for more than 14 days? Why is it still there?
  4. How many hours per week are you spending on work that a $15/hour VA could do? (email triage, scheduling, admin)

Brainstorm prompts

  • What would change next Monday if you had a real CEO beside you?
  • Which of your last 10 decisions do you regret? What pattern links them?
  • What are you doing that is comfortable but not important? List 3.

#2. Connect MCPs and tools

Install these MCP servers (or their equivalent integrations) to make the CEO Kit 10× more useful. Not all are required — pick the top 1–2 to start.

Notion

Decision log, board-update drafts, strategic one-pagers. The CEO Kit writes here.

Docs
Linear

OKRs and roadmap state. The skills pull current quarterly goals to anchor weekly planning.

Docs
Google Calendar

The weekly-priorities skill reads your week to spot where time is actually going vs. what you say matters.

Docs
Gmail

board-update skill drafts the email directly. Optional — you can paste manually.

Docs
Slack

Pulls recent channel context when writing investor updates — 'what shipped this week' without you typing it.

Docs
Don't have MCP set up yet?
MCP is the Model Context Protocol — Anthropic's standard for letting AI tools read from your software. Install guides: Claude Code MCP setup. ChatGPT uses Actions (OpenAPI) instead of MCP — same capabilities, different protocol.

#3. Define your cadence

The CEO Kit works because you run it consistently. Block the weekly ritual in your calendar now — even 15 minutes, non-negotiable, beats 2 hours every 3 months.

Daily

  • End of day·3 min
    One-line founder journal entry: what moved, what didn't, one thing to retry tomorrow.

Weekly

  • Monday 09:00·15 min/ceo-weekly
    Run the Monday planning ritual. Set 3 priorities for the week. Lock them.
  • Friday 17:00·10 minfounder-journal
    Retro: what happened vs. what was planned. Log one lesson.

Monthly

  • First Tuesday·30 minboard-update
    Draft the monthly investor update. 6 sections, short.

Ad-hoc

  • Before any major decision·20 minbuild-buy-kill
    Score the opportunity on Momentum / Market / Moat / Me. Get a verdict.
  • When you've already decided and want a challenge·10 minstrategic-advisor
    Spin up the strategic-advisor subagent to steelman + attack your reasoning.

#4. Use the skills

Every asset in the CEO Kit — with when to trigger it, the exact step-by-step, an example in/out, and the common pitfalls. Read them once; refer back as you run the cadence.

weekly-prioritiesSkill
Trigger
"plan my week" / "what should I focus on"
When to use
Every Monday morning. First thing before email.
Step-by-step
  1. Brain-dump everything on your mind (voice note works).
  2. The skill categorizes each item into MOVE / LEARN / MAINTAIN / DEFER / KILL.
  3. It cuts to 3 priorities, each with outcome, leading indicator, kill criteria.
  4. Output is a single markdown file you save in your decision log.
Example input
"I need to ship the onboarding flow, respond to the Y Combinator email, fix the Stripe webhook bug, hire a designer, write next month's investor update, and figure out pricing."
Example output
**Priority 1:** 20 new users completing onboarding in 7 days. **Priority 2:** Pricing test live by Friday. **Priority 3:** YC reply sent by Tuesday. Everything else → DEFER block on Friday afternoon.
Pitfalls
  • Don't ask for a 4th priority. The skill will refuse — this is a feature.
  • Don't set outcomes like 'ship feature X' — it'll push back. Outcome = observable result.
board-updateSkill
Trigger
"write the investor update" / "draft monthly board note"
When to use
First week of each month. Or the day before your next investor call.
Step-by-step
  1. The skill asks 6 questions: MRR + delta, cash + burn, ToFu metric, biggest win, biggest challenge, one specific ask.
  2. Fill each in 30 seconds — no polishing.
  3. Output is the 6-section investor email ready to send.
Example input
"MRR €4.2k → €5.1k. Cash €82k / €11k burn. Signups 180 → 240. Win: new activation flow doubled D7 retention. Challenge: hiring delayed — 3 senior eng candidates ghosted. Ask: intros to technical co-founders in EU."
Example output
A 400-word investor email with KPI table, 3 wins, 2 challenges, 1 specific ask, next-month plan. Ready to paste into Gmail.
Pitfalls
  • If the month was quiet, write a shorter update. The skill won't invent news — don't override it.
  • Asks must be specific. 'Any intros help' → the skill will refuse and ask for a role + company type.
build-buy-killSkill
Trigger
"should I build X" / "is this worth doing"
When to use
Any decision that takes more than 2 weeks or costs more than €500.
Step-by-step
  1. State the opportunity in one sentence.
  2. The skill scores 0–5 on Momentum / Market / Moat / Me, each with one line of evidence you provide.
  3. Total /20 → verdict: BUILD (≥14), BUY (10–13), DEFER (6–9), KILL (≤5).
  4. Output includes the first concrete action or the kill rationale.
Example input
"Should I build an affiliate program?"
Example output
Momentum 3 (no users asked). Market 2 (≤5% of MRR). Moat 1 (anyone can copy). Me 2 (≥2 weeks). Total 8/20 → DEFER. Re-evaluate when 3+ customers ask.
Pitfalls
  • Be honest on the Me score — most founders inflate it because they *want* to build it.
  • Momentum ≠ 'I think it's a big market.' Momentum = real signal in the last 14 days.
strategic-advisorSubagent
Trigger
Explicitly invoke: 'call the strategic-advisor' or via subagent menu.
When to use
You've already made a decision internally and want a hard challenge, not a plan.
Step-by-step
  1. Describe the decision in 3 sentences.
  2. Subagent steelmans your decision, then attacks it with 2–3 specific counter-arguments.
  3. Ends with one of: proceed with mitigations, modify, or pause + run a test.
Example input
"I'm raising a pre-seed round of €500k at a €3M cap."
Example output
Steelman: cap leaves room for seed. Attack: €3M cap + €500k = 14% dilution — you'll hit 40% founder dilution by Series A. Recommendation: modify. Raise €350k at €4M, extend runway via revenue. Test first: 2 commitments at the new terms before finalizing.
Pitfalls
  • This is not a planner. Don't ask it to write a plan.
  • Don't use it for reversible calls. Reserve for decisions >€5k or >1 month commitment.
/ceo:weeklySlash command
Trigger
Type /ceo:weekly in Claude Code.
When to use
Monday morning ritual. Non-negotiable.
Step-by-step
  1. Reads last week's entry from founder-log/weekly/.
  2. Asks for 60-second brain dump.
  3. Applies the weekly-priorities skill.
  4. Writes founder-log/weekly/YYYY-WW.md with 3 priorities locked.
  5. Ends with one provocative question.
Example input
(Just type /ceo:weekly.)
Example output
A file founder-log/weekly/2026-16.md with priorities, kill criteria, and Friday retro pre-filled.
Pitfalls
  • Don't skip the brain dump — that's where shadow backlog surfaces.
  • Cap at 15 minutes.
decision-logSkill
Trigger
"log this decision" / "I'm deciding X"
When to use
Decisions >€500, >1 week, cross-function, or hard to reverse.
Step-by-step
  1. Required fields: title, question, decision (imperative), ≥2 alternatives, one negative consequence (forced), observable 'I was wrong if' trigger, review date.
  2. Save to founder-log/decisions/YYYY-MM-DD-<slug>.md and append to index.
  3. Quarterly review checks 'wrong-if' signals.
Example input
"Switching to monthly + annual pricing"
Example output
decisions/2026-04-19-add-annual-plan.md with decision, 2 rejected alternatives, negative (cash-flow cliff if >40% take annual), wrong-if: annual <10% of new at 30d.
Pitfalls
  • Vague reasons → force concrete.
  • Non-observable revisit triggers → force numbers or dates.
founder-journalSkill
Trigger
"Friday reflection" / "retro the week"
When to use
Every Friday, 10 min, non-negotiable.
Step-by-step
  1. 3 specific wins (push back on 1 or 5).
  2. 1 named failure (no euphemism).
  3. 1 testable lesson.
  4. 1 Monday recalibration.
  5. 3-line energy check.
  6. Appends to founder-log/weekly/YYYY-WW.md.
Example input
(Friday prompt)
Example output
Wins (20 users, Stripe webhook fixed, investor intro). Failure (candidate ghosted). Lesson (require 20-min intro before any contract). Monday action (new hiring loop).
Pitfalls
  • 'No wins' → dig harder or that's the failure.
  • Skipping Friday entirely — 3 min > 0 min.
pitch-deckSkill
Trigger
"write my pitch" / "deck for [investor]"
When to use
Fundraise prep, specific investor, or update in deck form.
Step-by-step
  1. Pick variant: cold / warm / update.
  2. Strict 10 slides: title, why-now, problem, solution, how, traction, model, market, team, ask.
  3. One idea per slide. One metric on slides 2-6.
Example input
"cold deck, generic seed"
Example output
10 slides with raise target in title, why-now with data point, problem + customer quote, product shot, traction MoM, bottom-up market, founder-market fit, specific raise + 3 milestones.
Pitfalls
  • Team slide before traction → demote.
  • TAM=SAM=SOM chart → delete.
  • Competitive matrix → show wedge instead.
Your first win
Run /ceo:weekly next Monday. If your 3 priorities don't feel tighter than last week's, ask for a refund.